DESCRIPTION: Recent studies indicating that cytotoxic T cells (CTL) play a critical role in protective immunity to Mycobacterium tuberculosis suggest that an effective tuberculosis vaccine should contain epitopes that elicit such CTL. However, the mycobacterial epitopes that sensitize and serve as targets for CTL are largely unknown. We propose to analyze M. tuberculosis specific CTL generated in vitro and in vivo for the purpose of identifying the immunogenic epitopes of this organism. To generate primary CTL we will take advantage of a recently developed in vitro vaccination system, which utilizes peripheral blood dendritic cells (DC) to present antigens to naive CTL precursor cells. This system has been used successfully to generate HLA class I restricted CTL to a variety of antigens, including keyhole limpet hemocyanin and Human Immunodeficiency Virus derived peptides. Using this system in combination with peptide binding analysis, we will identify epitopes of the 19, 38, and 65 kDa proteins of M. tuberculosis that give rise to antigen specific CTL. We will also determine if the epitopes that elicit mycobacterium specific CTL, in vitro, serve as targets for CTL generated in vivo in individuals infected with tubercle bacilli or vaccinated with BCG. Although our initial studies will focus on CD8+, class I MHC restricted CTL from individuals expressing the HLA-A2 antigen (the most common HLA class I allele), our studies will eventually encompass CD4+ and CD8+ CTL from individuals with multiple HLA types. In addition, we will determine whether M. tuberculosis specific CTL affect the survival of tubercle bacilli in infected macrophages, and we will determine the spectrum of cytokines produced by these CTL. The results of these studies should facilitate the design of an effective subunit vaccine for use both in the prevention and treatment of tuberculosis.